17 Ocak 2012 Salı
14 Ocak 2012 Cumartesi
Computerized language translation started 58 years ago with IBM and Georgetown University
In 1954 IBM and Georgetown University teamed to produce the first English-to-Russian language computer translation program.
Hard to imagine but it has been 58 years since IBM and Georgetown University teamed up to run what they said was at the time the first English-to-Russian language computer translation program.
Perhaps even more interesting is that the individual phrases they were plugged into punch cards and run on the big IBM 701 mainframe in 1954, can now be typed into Google Translate on your smartphone and handled in about 10 seconds.
MORE INTERESTING STUFF: The weirdest, wackiest and coolest sci/tech stories of 2011
Still, the IBM program at the time was heralded as a major breakthrough. From an IBM press release on the event, which occurred on Jan. 7, 1954:
"A girl who didn't understand a word of the language of the Soviets punched out the Russian messages on IBM punch cards. The [IBM 701 which as known as the 'brain'] dashed off its English translations on an automatic printer at the breakneck speed of two and a half lines per second.
"'Mi pyeryedayem mislyi posryedstvom ryechyi,' the girl punched. And the 701 responded: 'We transmit thoughts by means of speech.'
"'Vyelyichyina ugla opryedyelyayetsya otnoshyenyiyem dlyini dugi k radyiusu,' the punch rattled. The 'brain' came back: 'Magnitude of angle is determined by the relation of length of arc to radius.'
"'Myezhdunarodnoye ponyimanyiye yavlyayetsya vazhnim faktorom v ryeshyenyiyi polyityichyeskix voprosov,' the girl tapped out. And the computer translated: 'International understanding constitutes an important factor in decision of political questions.'
"More than sixty Russian sentences were given to the 'brain' altogether. This amazing instrument was interrupted in its 16-hour-a-day schedule of solving problems in nuclear physics, rocket trajectories, weather forecasting and other mathematical wizardry. Its attention was turned at brief intervals from these lightning-like numerical calculations to the altogether different consideration of logic in an entirely new and strange realm for giant electronic data processing machines: the study of human behavior -- specifically, the human use of words. The result, as publicly proved today, was an unqualified success.
"Although IBM emphasized that it is not yet possible 'to insert a Russian book at one end and come out with an English book at the other,' [IBM] predicted that 'five, perhaps three years hence, interlingual meaning conversion by electronic process in important functional areas of several languages may well be an accomplished fact.'"
Interestingly, that sort of programming translation, while a hot topic during this period of time, proved difficult, expensive and ultimately controversial. In 1964 a group of scientists convened by the Department of Defense and National Science Foundation, known as the Automatic Language Processing Advisory Committee (ALPAC), evaluated language translation computing technology and largely killed the effort with a report issued in 1966.
The 1966 ALPAC report, "Language and Machines: Computers in Translation and Linguistics," according to a Wikipedia entry, "was highly critical of the existing efforts, demonstrating that the systems were no faster than human translations, while also demonstrating that the supposed lack of translators was in fact a surplus, and as a result of supply and demand issues, human translation was relatively inexpensive -- about $6 per 1,000 words. "
Longtime machine translation expert John Hutchins wrote years later: "The best known event in the history of machine translation is without doubt the publication of the report by the ALPAC in 1966. Its effect was to bring to an end the substantial funding of machine translation (MT) research in the United States for some twenty years. More significantly, perhaps, was the clear message to the general public and the rest of the scientific community that MT was hopeless. For years afterwards, an interest in MT was something to keep quiet about; it was almost shameful. To this day, the 'failure' of MT is still repeated by many as an indisputable fact. The impact of ALPAC is undeniable. Such was the notoriety of its report that from time to time in the next decades researchers would discuss among themselves whether "another ALPAC" might not be inflicted upon MT."
Beyond the controversy, computer language learning is ongoing today in some interesting ways.
networkworld
Çin’de 1912 tarihli Kur’an tercümesi
Çin’de 1912 yılından kalma Kur’an-ı Kerim tercümesi keşfedildi.
İslami Kültür araştırmacıları Çin’in Kuzeybatısı’ndaki Gansu ilinde, Çince yazılmış en eski Kur’an-ı Kerim nüshasını bulduklarını açıkladı. Kur’an-ı Kerim’in Çince manalarının yazılı olduğu bu el yazması mushafın yazılış tarihi 1912.
Lanzhou Üniversitesi’ndeki İslami Kültür Enstitüsü’nde yer alan eski kayıtlar arasında varlığı keşfedilen Kur’an tercümesinin Gansu Eyaleti’nin başkenti Lanzhou’da yaşamış iki ünlü imam ve Arap dili hattatı Sha Chong ve Ma Fu Lu’dan önce gerçekleştirildiği tahmin ediliyor. Enstitünün başkanı Deng Shi Ren’in ifadesine göre Sha ve Ma, Kur’an-ı Kerim manalarının tercümesine 1909 yılında başlayıp çalışmalarını 1912 yılında bitirdiler. Daha sonra Sha Çince nüshayı üç kitapta kopyaladı. Bu kopyalar başkent Lanzhou’da geniş çaplı olarak kullanıldı.
Deng Gansu’da 20. yüzyılda Kur’an-ı Kerim’in manalarının iki ayrı tercümesinin daha yapıldığını açıkladı. Ancak Sha ve Ma’nın tercümelerinin –Çince metnin bazı bölümlerinde Lanzhou lehçesi kullanılmasına karşın- Arapça’yla en uyumlusu olduğuna dikkat çekti.
Uzmanlar İslam’ın Çin’e Tang Hanedanı İmparatorluğu (618 MS – 907 MS) döneminde girdiğini söylüyor. Ancak o devirde yaşayan Çinli alimler tercüme ve tefsirde hata yapma korkusuyla Kur’an-ı Kerim manalarının tercümelerini yapmaya kalkışmamış.
1912 yılında gerçekleştirilmiş tercüme nüshasının bulunmasından önce uzmanlar Çince’ye çevrilmiş en eski Kur’an-ı Kerim nüshasının 1927 yılında Pekin’de yayınlanan nüshanın olduğunu tahmin ediyordu.
Timetürk
12 Ocak 2012 Perşembe
Ölünce internet hesaplarınız ne olacak?
"Değerli yeğenime, internette poker ve bingo hesaplarıma erişim hakkını bırakıyorum. Büyük kuzenim de tüm iTunes kredilerimi alsın."
Dijital mirasınızı kime bırakacağınızı düşündünüz mü?
İnanması zor gelebilir ama dijital ortamdaki varlıklarımız her geçen gün artarken bunlara ilişkin bir vasiyet de düzenlememiz gerekebilir.
Avukat Matthew Strain, müşterilerine şimdiden dijital mirasları konusunda tavsiyelerde bulunuyor ve vasiyetnamelere ekler düzenliyor.
Strain, "her geçen gün daha fazla miktarda fotoğraf, müzik ve kitap internet ortamında depolanır oldu. Tümü dijital formatta. Dolayısıyla insanlar öldükten sonra bunlara ne olacağı sorusu da giderek daha fazla önem kazanıyor" diyor.
Zira internetteki kimi dijital fotoğraf ya da videoların, uygulamaların duygusal olduğu kadar parasal değeri bulunduğu da muhakkak.
Dijital mal varlıklarının sorumluluğunu üstlenmeyenlerin öldüklerinde bu varlıkları kaybetme hatta akrabalarına ödenmemiş fatura bırakması dahi mümkün.
iCroak adlı internet sitesinde dijital malvarlıklarının öldükten sonra idare edilmesini isteyenlere özel hizmet veriliyor.
Yılda 10 ila 15 İngiliz sterlini, yani 30 – 45 Türk lirası ya da tek seferlik 150 sterlin yani 430 Türk lirası karşılığı kullanıcılar malvarlıklarını kategorize ederek "koruyucu" ya da "vasi" hesaplar oluşturabiliyor.
Koruyucu, vasi seçilen kişiye özel kullanıcı adı ve şifresi gönderiliyor.
Bu kişi, dijital mirası ancak kendisini bu göreve atayan kişinin ölüm belgesi doğrulandığında görebilecek.
Kuşkusuz kimilerinin ileride kimsenin görmesini istemeyeceği ya da onları utandırabilecek hesapları olabilir, iCroak bunun için de bir seçenek sunuyor ve öldüğünüzde tüm bilgilerinizin silinmesini sağlayabiliyor.
Goldsmiths Üniversitesi'nin yürüttüğü bir araştırmaya göre İngilizler, 2,3 milyar sterlin değerinde müzik, film, uygulama ya da internet üyeliğine sahip.
Aynı araştırma her 10 kişiden birinin şifrelerini vasiyetlerine eklediğini, yani yakınlarının dijital hazinelerine ulaşmasını istediğini gösteriyor.
Twitter Google'ın sosyal medya arama motorundan şikayetçi
Twitter, Google'ın, arama sonuçları içinde, kendi sosyal paylaşım sitesi Google+'dan sonuçlara ön sırada yer vermesinden rahatsız olduğunu açıkladı.
Sosyal paylaşım sitesinde paylaşılanların Google arama sayfasındaki sonuçlar arasında yer alması bu sitelere olan trafiği artırıyor.
Sosyal paylaşım ve mikro blog sitesi Twitter'ın avukatlarından Alex Macgillivray, Twitter'da paylaştığı mesajında, "Bugün internet için kötü bir gün" dedi.
İnternet arama motoru devi Google, Facebook'a rakip sosyal paylaşım sitesini güçlendirmek için yoğun çaba harcıyor.
Google+'da paylaşılanların özel olarak hazırlanan Google arama sayfasında yer bulacak olması, bu siteye olan talebi artırabilir.
Twitter'ın şikâyetine yanıt veren Google ise, değişikliklerin, internet aramasının giderek kişiselleştirilmesinin bir parçası olduğunu söyledi.
Google, arama sonuçlarında, bir süredir kişiye özel sonuçlar sergilemeye başladıklarını da ifade ediyor.
Google'ın yaptığı değişikler sayesinde, kullanıcılar Google+ fotoğrafları, iletileri ve durum raporları arasında da arama yapabilecek.
Google+'ya üye kişilerin profillerinde de arama yapılabilecek.
Konuyla ilgili Twitter'daki sayfasından yorumlar yapan Twitter'ın avukatı Macgillivray, geçmişte Google için çalışıyordu.
Twitter da yaptığı resmi açıklamada, Macgillivray'in şikâyetlerine benzer noktaları tekrarladı.
Yunan ve İspanyollar Almanca öğrenmek için sırada
Euro bölgesindeki borç krizinden en az etkilenen ülkelerden olan Almanya, ekonomik gücü ile kültürel cazibesini de arttırır görünüyor.
Alman dil ve kültürünü ve dilini öğreten Goethe Enstitüsü'ne göre Almanca öğrenenler hızla artıyor.
Enstitünün Frankfurt'taki merkezinde 20 yılı aşkın süredir ders veren Günther Schwinn-Zur, işlerin hiç şimdiki kadar iyi olmadığını söyledi.
Schwinn-Zur, geçen yıl derslerine talebin yüzde 30'un üzerinde artış kaydettiğini söyledi.
Benzer bir eğilim ülkenin geri kalanındaki şubelerde de görülürken, yeni öğrenciler arasında en büyük grubu Yunan ve İspanyolların oluşturduğu bildiriliyor.
Financial Times gazetesine göre bu durumun temelinde Alman ekonomisine ilişkin yaygın olumlu izlenim yatıyor.
Federal istatistik idaresi verilerine göre Alman ekonomisi 2011'de yüzde 3 büyüdü.
Bu oran Amerika Birleşik Devletleri ya da Euro Bölgesi'nin tamamında görülenin iki katı.
Dolayısıyla genç işgücü için Almanya cazip bir hedef durumunda.
Bununla birlikte dün açıklanan veriler Almanya'da büyümenin hız kesmeye başladığını göstermişti.
Yılın son üç ayında eksiye geçilmiş olabileceği düşünülürken, yılın ilk üç ayında da benzer bir durum yaşanmasından endişe ediliyor.
-BBC-
11 Ocak 2012 Çarşamba
The man who wants to translate the Web
(CNN) -- I want to translate the Web into every major language: every webpage, every video, and, yes, even Justin Bieber's tweets.
With its content split up into hundreds of languages -- and with over 50% of it in English -- most of the Web is inaccessible to most people in the world. This problem is pressing, now more than ever, with millions of people from China, Russia, Latin America and other quickly developing regions entering the Web. In this TED talk, I introduce my new project, called Duolingo, which aims at breaking the language barrier, and thus making the Web truly "world wide."
We have all seen how systems such as Google Translate are improving every day at translating the gist of things written in other languages. Unfortunately, they are not yet accurate enough for my purpose: Even when what they spit out is intelligible, it's so badly written that I can't read more than a few lines before getting a headache. This is why you don't see machine-translated articles on CNN.
With Duolingo, our goal is to encourage people, like you and me, to translate the Web into their native languages.
Now, with billions and billions of pages on the Web, this can't be done with just a few volunteers, nor can we afford to pay professional translators. When Severin Hacker and I started Duolingo, we realized we needed a way to entice millions of people to help translate the Web. However, coordinating millions of contributors to translate language presents two major hurdles. First, finding enough people who are bilingual enough to help with translation is difficult. Second, motivating them to do it for free makes this next to impossible.
The idea behind Duolingo is to kill two birds with one stone by solving both of these problems simultaneously. We accomplish this by transforming language translation into something that anyone can do -- not just bilinguals -- and that millions of people want to do: learning a foreign language.
It is estimated that over one billion people worldwide are learning a foreign language, with millions doing so using computer programs. With Duolingo, people learn a foreign language while simultaneously translating text.
When you learn on Duolingo, the website gives you exercises tailored specifically to you that teach you every aspect of the new language. You may be asked to translate a sentence, to pronounce or listen to a phrase, or to describe what you see in an image.
Some of the sentences you translate come from real websites. By having multiple students translate each sentence, and then choosing the best one, Duolingo produces translations that are as accurate as those from professional language translators.
Because you create valuable translations as a side effect, learning on Duolingo is 100% free: no ads, no hidden fees, no subscriptions. Duolingo entails a new business model that allows anyone online, regardless of socioeconomic status, to have access to education.
For example, the leading language-learning software sells for over $500, which is beyond the means of the majority of the world's population. If language education is offered free of charge in exchange for students' performing useful tasks, those who cannot afford to pay with money pay with their time -- time that would have been spent learning anyway.
This is how I want to translate the Web. Now go on and sign up for Duolingo.
With its content split up into hundreds of languages -- and with over 50% of it in English -- most of the Web is inaccessible to most people in the world. This problem is pressing, now more than ever, with millions of people from China, Russia, Latin America and other quickly developing regions entering the Web. In this TED talk, I introduce my new project, called Duolingo, which aims at breaking the language barrier, and thus making the Web truly "world wide."
We have all seen how systems such as Google Translate are improving every day at translating the gist of things written in other languages. Unfortunately, they are not yet accurate enough for my purpose: Even when what they spit out is intelligible, it's so badly written that I can't read more than a few lines before getting a headache. This is why you don't see machine-translated articles on CNN.
With Duolingo, our goal is to encourage people, like you and me, to translate the Web into their native languages.
Now, with billions and billions of pages on the Web, this can't be done with just a few volunteers, nor can we afford to pay professional translators. When Severin Hacker and I started Duolingo, we realized we needed a way to entice millions of people to help translate the Web. However, coordinating millions of contributors to translate language presents two major hurdles. First, finding enough people who are bilingual enough to help with translation is difficult. Second, motivating them to do it for free makes this next to impossible.
The idea behind Duolingo is to kill two birds with one stone by solving both of these problems simultaneously. We accomplish this by transforming language translation into something that anyone can do -- not just bilinguals -- and that millions of people want to do: learning a foreign language.
It is estimated that over one billion people worldwide are learning a foreign language, with millions doing so using computer programs. With Duolingo, people learn a foreign language while simultaneously translating text.
When you learn on Duolingo, the website gives you exercises tailored specifically to you that teach you every aspect of the new language. You may be asked to translate a sentence, to pronounce or listen to a phrase, or to describe what you see in an image.
Some of the sentences you translate come from real websites. By having multiple students translate each sentence, and then choosing the best one, Duolingo produces translations that are as accurate as those from professional language translators.
Because you create valuable translations as a side effect, learning on Duolingo is 100% free: no ads, no hidden fees, no subscriptions. Duolingo entails a new business model that allows anyone online, regardless of socioeconomic status, to have access to education.
For example, the leading language-learning software sells for over $500, which is beyond the means of the majority of the world's population. If language education is offered free of charge in exchange for students' performing useful tasks, those who cannot afford to pay with money pay with their time -- time that would have been spent learning anyway.
This is how I want to translate the Web. Now go on and sign up for Duolingo.
Google’s ‘Babel fish’ heralds future of translation
In Douglas Adams’s famous Hitchhiker’s Guide to the Galaxy series of science-fiction books, interstellar species use Babel fish — “small, yellow, leech-like” creatures that feed on “brain-wave energy” — to translate speech in real time.
A team of developers at Google is working on the real thing, using statistical models to translate different languages, including Afrikaans, on the Web and on mobile phones, using voice input and output as well as text.
TechCentral sat down with Google Translate research scientist Ashish Venugopal at Google’s headquarters in Silicon Valley last week and asked him about the stumbling blocks to effective real-time translation and the future of the technology. This is an edited transcript of that interview.
TechCentral: How many languages does Google Translate now support?
Ashish Venugopal: There are 63 languages supported. That’s a lot of languages. How do we get all that data in there? If we tried manually to give the system those languages, it would be a hopeless task. The only possible way we could do this is to harness the power of machine computation. We build statistical models that are automatically training themselves and learning all the time. As people translate new content on the Web, our systems pick this up and it adds the words. The system is constantly reading and analysing the Web. It’s a statistical approach. The idea is that once we learn the essential model of how to speak a word, and we can apply that to every word. We haven’t memorised every word.
Are there any difficult languages that make it hard to get translation right?
Yes, there are some incredibly tough languages. If your language is very different from English, for example, then it will be very difficult to translate it to English. We use English as an intermediate language and so if you were translating from Russian to Japanese, we’d translate the Russian to English and then to Japanese.
When we talk about a “tough” language, it’s one that is really different compared to English. There are languages that are very different in multiple dimensions.
The first question to ask is, what is the order of words from in one language compared to English. In English, we’d put the subject first, then the verb and then the object, whereas the Japanese have the subject first, then the object and finally the verb. We have to teach computers how to recognise this reordering pattern.
We don’t tell the computer how to translate every sentence. We give it general patterns to look for. When it sees new data, it uses those patterns, matches that to data and then comes up with a model that it uses to translate sentences.
When we say languages are harder, they’re harder because of the ordering of words, they’re harder because there may be different notions of what a word even is. In English, you say you put the phone on the table — “phone” and “table” are objects and “on” is an additional word that explains what’s happening. In other languages, the “on” could be glued onto the word “phone” or “table” and we have to teach the computer that “on” could be connected to the object or be separate from it.
All these issues get easier when there’s more source data. We launch languages when we feel they are adding value to somebody. We have “alpha” or experimental languages where we were just able to launch the system, as opposed to it being fluent and correct. The alpha languages tend to have less source data available online.
What are the main stumbling blocks to this technology and what will be possible in the future?
We are really reliant on the source data. The first stumbling block for a new language is, is there data on the Web? Once there’s enough content on the Web and as we build our system … on average it works really well. On average, you’ll be very impressed with it. But every once in a while you’ll be irritated with it.
Because of the statistical approach, you may enter something and get some crazy translation. What we are trying to do is limit those crazy translations and ensure in all cases we are providing a reasonable translation.
This really comes from the fact that this is a statistical system. We’ve built it so you can literally put anything into it. We will translate anything you give us. It might be good or it might be bad, but on average it will be quite impressive.
What we are really working on now is clipping the bottom end of the cases where we make mistakes. We see these issues in languages that are very different compared to English. Russian, for example, adds a lot of information to words and they get longer and longer and when we translate we sometimes make mistakes there.
In the future, in a reasonably short time, we will take machine translation for granted, as part of our everyday lives. I mean that from an 80-20 standpoint, where 80% of the use cases we’ll be able to address effectively. The last 20% will be incredibly hard. That speaks to the fact that machine translation won’t be a substitute for a human translator.
No one is going to take an important political speech and put it into machine translation to publish it in 20 different languages. Our goal is not to create artificial intelligence; our goal is to provide an 80% solution where you’ll be able to understand the political speech’s point, but not it’s rhetoric, not it’s beauty necessarily.
Is the future of this technology instant voice translation using devices like mobile phones to facilitate real-time translation of conversations?
We can do that already, but not simultaneously. It’s not an immediate goal. It’s a matter of where we are focusing now. There’s still more work to be done on the quality side before we can start to develop this continuous form of operation.
Will you continue to do translation in the cloud (online on servers) or will it move down to devices like phones as they get more powerful?
We make all our decisions purely based on quality. We want to ensure the highest quality translations are delivered to our users in the shortest possible time and that’s leaning towards the cloud for now, but that might change.
What sort of computing power does Google Translate require?
We use the full power of Google’s search engine. The reason Google Translate exists is because of the investments made in search. We sit on top that search infrastructure.
Do you have a team of linguists working all over the world?
We have a team of statisticians, all working right over there [points and laughs]. It’s less linguistically orientated. There are linguistic ideas that influence our decisions. To give you an example, when I was working on the last set of Indian languages that were launched, I didn’t use any linguistic knowledge; I used Wikipedia and my grandmother. So, it’s Wikipedia, my grandmother and statistics. That’s what we use to put a language together. — Duncan McLeod, TechCentral
A team of developers at Google is working on the real thing, using statistical models to translate different languages, including Afrikaans, on the Web and on mobile phones, using voice input and output as well as text.
TechCentral sat down with Google Translate research scientist Ashish Venugopal at Google’s headquarters in Silicon Valley last week and asked him about the stumbling blocks to effective real-time translation and the future of the technology. This is an edited transcript of that interview.
TechCentral: How many languages does Google Translate now support?
Ashish Venugopal: There are 63 languages supported. That’s a lot of languages. How do we get all that data in there? If we tried manually to give the system those languages, it would be a hopeless task. The only possible way we could do this is to harness the power of machine computation. We build statistical models that are automatically training themselves and learning all the time. As people translate new content on the Web, our systems pick this up and it adds the words. The system is constantly reading and analysing the Web. It’s a statistical approach. The idea is that once we learn the essential model of how to speak a word, and we can apply that to every word. We haven’t memorised every word.
Are there any difficult languages that make it hard to get translation right?
Yes, there are some incredibly tough languages. If your language is very different from English, for example, then it will be very difficult to translate it to English. We use English as an intermediate language and so if you were translating from Russian to Japanese, we’d translate the Russian to English and then to Japanese.
When we talk about a “tough” language, it’s one that is really different compared to English. There are languages that are very different in multiple dimensions.
The first question to ask is, what is the order of words from in one language compared to English. In English, we’d put the subject first, then the verb and then the object, whereas the Japanese have the subject first, then the object and finally the verb. We have to teach computers how to recognise this reordering pattern.
We don’t tell the computer how to translate every sentence. We give it general patterns to look for. When it sees new data, it uses those patterns, matches that to data and then comes up with a model that it uses to translate sentences.
When we say languages are harder, they’re harder because of the ordering of words, they’re harder because there may be different notions of what a word even is. In English, you say you put the phone on the table — “phone” and “table” are objects and “on” is an additional word that explains what’s happening. In other languages, the “on” could be glued onto the word “phone” or “table” and we have to teach the computer that “on” could be connected to the object or be separate from it.
All these issues get easier when there’s more source data. We launch languages when we feel they are adding value to somebody. We have “alpha” or experimental languages where we were just able to launch the system, as opposed to it being fluent and correct. The alpha languages tend to have less source data available online.
What are the main stumbling blocks to this technology and what will be possible in the future?
We are really reliant on the source data. The first stumbling block for a new language is, is there data on the Web? Once there’s enough content on the Web and as we build our system … on average it works really well. On average, you’ll be very impressed with it. But every once in a while you’ll be irritated with it.
Because of the statistical approach, you may enter something and get some crazy translation. What we are trying to do is limit those crazy translations and ensure in all cases we are providing a reasonable translation.
This really comes from the fact that this is a statistical system. We’ve built it so you can literally put anything into it. We will translate anything you give us. It might be good or it might be bad, but on average it will be quite impressive.
What we are really working on now is clipping the bottom end of the cases where we make mistakes. We see these issues in languages that are very different compared to English. Russian, for example, adds a lot of information to words and they get longer and longer and when we translate we sometimes make mistakes there.
In the future, in a reasonably short time, we will take machine translation for granted, as part of our everyday lives. I mean that from an 80-20 standpoint, where 80% of the use cases we’ll be able to address effectively. The last 20% will be incredibly hard. That speaks to the fact that machine translation won’t be a substitute for a human translator.
No one is going to take an important political speech and put it into machine translation to publish it in 20 different languages. Our goal is not to create artificial intelligence; our goal is to provide an 80% solution where you’ll be able to understand the political speech’s point, but not it’s rhetoric, not it’s beauty necessarily.
Is the future of this technology instant voice translation using devices like mobile phones to facilitate real-time translation of conversations?
We can do that already, but not simultaneously. It’s not an immediate goal. It’s a matter of where we are focusing now. There’s still more work to be done on the quality side before we can start to develop this continuous form of operation.
Will you continue to do translation in the cloud (online on servers) or will it move down to devices like phones as they get more powerful?
We make all our decisions purely based on quality. We want to ensure the highest quality translations are delivered to our users in the shortest possible time and that’s leaning towards the cloud for now, but that might change.
What sort of computing power does Google Translate require?
We use the full power of Google’s search engine. The reason Google Translate exists is because of the investments made in search. We sit on top that search infrastructure.
Do you have a team of linguists working all over the world?
We have a team of statisticians, all working right over there [points and laughs]. It’s less linguistically orientated. There are linguistic ideas that influence our decisions. To give you an example, when I was working on the last set of Indian languages that were launched, I didn’t use any linguistic knowledge; I used Wikipedia and my grandmother. So, it’s Wikipedia, my grandmother and statistics. That’s what we use to put a language together. — Duncan McLeod, TechCentral
5 Ocak 2012 Perşembe
Google’da tercüme et ‘soykırım’ desin
Google’ın çeviri motoru küçük harfle “ermeni soykırımı yoktur” cümlesini diğer dillere “vardır” diye çevirdi!!
Fransa Parlamentosu’nun “Ermeni soykırımı yoktur” diyene 1 yıl hapis ve 45 bin Euro para cezasını öngören yasa taslağını kabul etmesi ile ortaya çıkan gerginlik, dünyanın en büyük arama motoru Google’a da sıçradı. Google’ın çeviri servisi Google Translate’in soykırımla ilgili çevirisi ‘teknik hata’ verdi.
translate.google.com adresinin Türkçe’den diğer dillere çeviri bölümünde “Ermeni soykırımı yoktur” cümlesi “Ermeni soykırımı yoktur” şeklinde çevrilirken, Ermeni kelimesindeki ‘e’ küçük harfle yazıldığında, İngilizce karşılığında “There is the Armenian genocide” yani “Ermeni soykırımı vardır” çevirisi yapıldığı tespit edildi. Almanca ve İspanyolca çeviride de aynı sonuç çıktı.
Ermeni kelimesi küçük harfle başladığı anda “Soykırım yoktur”un, “Soykırım vardır” diye çevirilmesi üzerine HABERTÜRK olarak konuyla ilgili Google’in Türkiye’deki temsilcileri ile temasa geçtik. Hatanın teknik bir sorundan kaynaklandığını belirten Google yetkilileri, hatanın siyasi veya tarihi bir tartışmadan kaynaklanmasının sözkonusu olmadığı belirtirken, çevirilerin “insan müdahalesi olmadan, en gelişmiş teknolojiyle” yapıldığı ifade etti.HABERTÜRK’ün uyarısı üzerine düzeltme yönünde harekete geçen Google’da çeviri bölümündeki bu hata 1 saat içinde düzeltildi.
‘ÇEVİRİDE İNSAN MÜDAHALESİ YOK’
Konuyla ilgili olarak başvurduğumuz Google’dan yapılan açıklamada, “Google’ın çeviri teknolojisinin herhangi bir insan müdahalesi olmadan otomatik olarak ileri teknoloji ile çalıştığı” belirtildi. Açıklamada, “Google Translate, bir çeviri yaparken, yüz milyonlarca belge arasından yaptığı karşılaştırma sonrası sizin için en uygun çeviriyi verir ve bu çeviriler tam bir çeviri olmadığı gibi eksiklikler ve yanlışlıklar içerebilir. Eğer bir kişi doğru veya uygun olmayan bir çeviri ile karşılaştığını düşünüyorsa, bizi bu konuda bilgilendirmesi yoluyla söz konusu hatanın en kısa sürede giderilmesini sağlayabilir” denildi.
HT GAZETE
Veda Hutbesi'nde 'tercüme' krizi
Cami duvarına asılan ve "kadının hafifçe dövülebileceğine" vurgu yapan Veda Hutbesi, bir vatandaş tarafından Cumhurbaşkanı'na şikâyet edilince tartışma konusu oldu
Kadına şiddetin tartışıldığı bir dönemde, İstanbul Beylikdüzü'ndeki Fatih Sultan Mehmet Camisi'nin duvarına asılan Veda Hutbesi'nde "İtaat etmeyen kadını dövün" ifadesinin yer alması tartışmaya neden oldu. İbadete 2009'da açılan caminin giriş kapısının bir tarafına İstiklal Marşı, diğer tarafına ise Veda Hutbesi'nden bölümler işlendi. Cami duvarında asılı levhada da Veda Hutbesi'ne yer verildi. Hz. Muhammed'e ait olduğu ileri sürülen sözlerde, itaat etmeyen kadınların erkekler tarafından dövülebileceği ifadesi yer aldı. Beylikdüzü'nde yaşayan Cengiz Alaçayır adlı vatandaş, bu durumu Cumhurbaşkanı Abdullah Gül'e bir elektronik posta ile bildirdi. Cumhurbaşkanlığı da konuyu Diyanet İşleri Başkanlığı'na havale etti.
'TERCÜME HATASI'
Tartışmaların odağındaki Fatih Sultan Mehmet Camisi'nin Dernek Başkanı Hasan Çelebi, "Biz Diyanet İşleri'nin sitesinden alıp, noktasına virgülüne dokunmadan koyduk" dedi. Nisa suresinin 34. ayetinin dayanak gösterildiği meale ilahiyatçılar farklı yaklaştı. SABAH Yazarı Doç. Dr. Nihat Hatipoğlu, "'Cennet analarımızın ayakları altındadır' diyen peygamberimizin kadınlara yönelik böyle hutbe vermesi mümkün değil. Metinleri incelemek ve bu metinlerden bir sonuç çıkarmak daha doğru bir yöntem. Söylendiği dönemdeki uygulamayı iyi değerlendirmek gerekiyor. Ayrıca bu hutbenin cami duvarına asılması uygun değil" şeklinde konuştu. Ankara Üniversitesi İlahiyat Fakültesi Dekanı Prof. Dr. Nesimi Yazıcı ise: "Diyanet geçmişte yapılan bir tercüme hatası varsa düzeltmeli. Bana göre tercüme hatası" derken Din İşleri Yüksek Kurul Üyesi Prof. Dr. Bünyamin Erul da, "Veda hutbelerinin çevirilerinde hatalar yapılmış olabilir. Hayatı boyunca kadına el kaldırmamış peygamber efendimizin böyle bir şeyi direkt söylemesi mümkün değildir" ifadesini kullandı. Tercümeden kaynaklanan bir sorun olduğuna dikkat çeken eski Beykoz Müftüsü CHP İstanbul Milletvekili İhsan Özkes ise, "Dövmekten kasıt tedip etmek, hizaya getirmek olabilir" dedi.
İŞTE DUVARA İŞLENEN VEDA HUTBESİ
Camideki Hutbe'de "Sizin kadınlar üzerindeki hakkınız; yatağınızı hiç kimseye çiğnetmemeleri, hoşlanmadığınız kimseleri izniniz olmadıkça evlerinize almamalarıdır. Eğer gelmesine müsaade etmediğiniz bir kimseyi evinize alırlarsa, Allah, size onları yataklarında yalnız bırakmanıza ve daha olmazsa hafifçe dövüp sakındırmanıza izin vermiştir" ifadesi yer alıyor.
-Sabah-
Kaydol:
Kayıtlar (Atom)








